Hi all,
In making translational reporters of alternative splicing isoforms of icl-1 and C08F11.14 I hit on inequal exon boundaries between Wormbase/Aceview and UCSC. Which database is most trustworthy for this or how can I make sure I develop the right constructs?
Kind regards,
ASM
Hi ASM,
when you compare across various databases, it is important to check which version of WormBase they are based on.
AceView as example, is based on a WormBase import from December 2009 (most likely WS210), but they improved the gene model display.
The current release of WormBase is WS244, which you can see on the wormbase website.
UCSC is showing WS190, but due to technical reasons, the underlying sequences are called differently.
EnsEMBL and EnsEMBL metazoa currently show WS240.
While our influence on which data other sites show/use is limited, we are
a.) working together with EnsEMBL to include every 5th WormBase release (so WS245 will show up soon)
b.) working with UCSC to provide updated nematode data through their genome hubs (you can use http://www.ebi.ac.uk/~mh6/myHub/hub.txt to see my proof of concept based on WS244)
The general rule of thumb is: the newer the WS release, the better the curation.
But of course if you find something that looks like it might be wrong/needs curation/improvement, contact help@wormbase.org
Separate from the efforts of the curators, the raw data they work with is available. You can look at the available sequenced RNAs (ESTs, OSTs, etcetera) and at conservation with other Caenorhabditis nematodes, and make up your own mind.